So, the Lumenlab CNC kit is designed and finished! (Sure, like it could ever be finished...)

Anyway, this is a phenomenal machine for not very much money. The breakthrough was sourcing the recirculating linear ball bearings (yes!) in Asia. We can provide the entire set of bearings, mounts and rails to build this machine for ~$399.00! (That's the price on one bearing from Thompson, USA!) With our part-cut list, you can go to Onlinemetals.com, punch in the cuts you need and have the 6061 billet delivered already cut! (~$125) Add our complete motor and screw kit for another $300 (3 nema 34, one nema 23). All that's left is to drill and tap the mounting holes for our hardware, bolt it all together, and get yourself a T-Slot aluminum base for it all (we will also offer, no price yet..) I feel confident I can sell you the entire kit for $999! That is insane for this machine!! All you have to do is assemble it. That takes care of the mechanical side of the equation....

We also offer an all-in-one 19" touch screen computer that you can buy preprogrammed to run our machine, and that also contains all of the hardware drivers for the motors! Pricing for that is under $1k, but it is not yet ready. For now, you'll have to hack together your own PC and driver combo, but we can help!

Notable features:

*True 26" x 50" cutting area (yes!)

*Massive billet aluminum construction

*Very powerful, made for full size routers (or whatever you want to put on there, use your imagination!)

*Can cut on a bed flush with the top of the T-Slot, or remove the bed and cut beneath the framework...this means you can set the machine down on a surface, cut it, and move it as you need!

*I have designed the LLCNC with a very special capability: If you buy our inexpensive up-coming kit, you can cut 4 x 8 sheets with it (or longer)! This is because I invented a stepper driven feed-roller that can feed the sheet under the ways, which I left high enough for clearance up to 1.5". Yay! This puts the machine's capabilities on par with very very expensive pieces of equipment!

*8" of Z! Enough for 3d, sculpting and relief.

*Very adaptable, useful for lots besides routing and milling. We have a lot of neat things coming, like wax deposition printing for rapid prototyping. Laser cutters and more!

Pics: A few drawings.

This is actually not the same gantry as the final. We have made it yet simpler by eliminating the need to drill through the gantry-end for the X ways! All you have to do is drill and tap (or buy the complete kit from us).

Click to view attachment

These are self explanatory:

Click to view attachment

Click to view attachment

Very nice and clean design.

Appreciate it Joe! Means a lot from the "Guru".

That crackling sound is my keyboard short circuiting...

Very sexy! Especially when you see how much more you can get over the Craftsman for just a little more money and the computer with it.
The power, space and versatility of this is just terrific.

I think I may have just found my next hobby!


PS. This WILL come loaded with pre-programmed projects designed to increase the Wife Acceptance Factor, right?

My keyboard shorted out from all the drool.

When's Father's Day???

You'll never pick up a saw/drill/router again! I have made this machine so easy to use that you'll be cutting cabinets with filigrees for your wife on the first day. edit: HAHA I just noticed the Craftsman has a 14" cutting width??? USELESS!


Every day!

I do like the looks of this machine. I am not trying to be critical, but the simply supported beams will deflect a little. If the X and Y rails were supported it would be a lot better machine.

I have tried 3/4" steel rod spanning 36" and the deflection less than calculating a simply supported beam, but not by much.

deflection = (PL^3)/(48EI)
Where:
P = Load
E = Modulus of Elasticity (Young's Modulus)
I = Moment of Inertia

I tried to upload an Excel workbook, but no sucess...

Yea this will be a present to myself

Awesome work !!

The x axis is supported by 35mm ways! The y axis by 25mm, and the z by 20mm. It is HUGE rail. The only thing close to 3/4" we use is for 8" of z. The rail is hardened and chrome plated. Also if the rails were supported, we'd lose the "feed through" option, which is truly revolutionary for any CNC thus invented. We've been running a similar machine non-stop and have never observed any flex no matter how fast we push it. The bit breaks first!

Here's a pic I took tonight of the bearings...this is about 1/2 of what you'd get for $400!
Click to view attachment

Wow! 35 mm. I stand corrected. Looks like a quality kit.

*I have designed the LLCNC with a very special capability: If you buy our inexpensive up-coming kit, you can cut 4 x 8 sheets with it (or longer)! This is because I invented a stepper driven feed-roller that can feed the sheet under the ways, which I left high enough for clearance up to 1.5". Yay! This puts the machine's capabilities on par with very very expensive pieces of equipment!

So How Much floor space does the feed roller take up does it move north south east west? OY it sounds like a lot of error
so much moving around....Like me trying to dance the Lindy.

How would you compare this machine to say the mechmate....I like the price, looks nice.

This will be fine for light cuts in light material but unsupported rails will be the week link. Other designs with unsupported rails work with much smaller dimensions. deflection goes up greatly with beam length and 50" is very very long for an unsupported rail even if it is more than an inch thick. if you don't believe me play around with this calculator http://www.geocities.com/richgetze/.

Man those pieces look sexy! Does the hat come too?

This deflection business concerns me a little, what's considered "light material"? I'm guessing MDF would fall into that category and steel would be outside that category, but how about aluminum? Can the effect of deflection be lessoned by running the machine slower (or maybe by making multiple passes)?

With a span of 50" and 10 lbs on each rail, each rail would deflect 0.005" (0.126 mm) This assumes the Y axis is in the center of the X axis travel and the load is split even on each rail. The 20 lbs includes the weight of the Y, Z and router parts.

The tool forces would be added. The deflection caused by these forces is dependent on what speeds, feeds and the material you are cutting.

It all comes down to how much deflection you can live with. All machines have these problems... typically the more expensive ones are more rigid.

Do people ever take these deflections into account when making their pattern (I don't know the official term)? Like, if I've observed that the kerf of my saw is 1/8" I might move my cutting line 1/16".

The deflection would vary from middle to end of the rails due to the weight of the gantry components. I think this could be programmed out.

The deflections from the tool bit forces would be complicated, but I would guess it would be possible. I am not sure it would be practical.

I would guess that if you programmed for the weight deflection and kept your tool forces down with slower speeds and shallow cuts per pass you get better results.

It really comes down to what you are making. Gears and bearing housings need better precision than plaques.

Nonsense. We run a smaller version of this machine all day long. The X is 44" of 20mm rail and has never suffered from any deflection issues! We cut 1/8" aluminum sides and backs for the Q and have no issues regarding accuracy. The key to stability is the wide contact area of the gantry bearing surface, you can't measure deflection as per a fulcrum in this case. The LLCNC is for cabinetry, modeling, prototyping, enclosures, signage etc etc etc. If you want .0001 accuracy you need a mill.
Click to view attachment
Click to view attachment


Yep, free hat with first 25 orders.

The wide base helps a little but you still have to deal with the deflection. There are formulas in the Machinery's Handbook that describe the deflection you will get. The numbers I gave in the above post were for 6 inch bearing spread on the gantry.

How are you measuring flatness from the work that you have done? 0.126 mm is not very much deflection on a large piece.

I am sure that the machine is quality for it's intended audience, but I am also sure there is a small amount of deflection with this design.

I can only dream of having such a machine - as I'm only able to rent an apartment at this present time. How I would love to build my own cabinets, furniture and, of course, DIY projectors using this machine.

brain, have you thought about making a video of the CNC machine in action and posting/advertising via youtube or some other video upload service?

Sure, I don't disagree with you, it was the other guy's "chicken little". There's deflection surely. Hell, everything on Earth is slightly curved! I'm just saying it's insignificant for the intended use! If you can't live with a few thou, get a mill.

The roller itself takes no room, as it attaches within the perimeter of the existing frame. The user sets up "roller work supports" based on their need. The Y axis is "parked" close to the roller, then Y is given over to the feed roller. As for error by the roller; the (pinch)roller and stepper work fine for things like signage, where very-high accuracy isn't needed. For more accuracy, use an "encoder strip" attached to the workpiece and relegate positioning to this encoder rather than just counting steps for Y.

As for Mechmate, I have no experience. It seems to be well designed, but if you want to cut 4x8, you'd need a big machine!

Hey guys. Just thought I'd chime in.
Grayson (brainchild) and I have spent the past several months designing this CNC machine. Hope you like the drawings, every bolt and screw thread was drawn by scratch on our Q computer.

Programmatic Requirements (aka design goals):
-Robustness (made of a reasonably hard alloy of aluminum, 6061, that is easy to machine)
-Inexpensive, but not cheap.
-High enough accuracy to perform the vast majority of required CNC operations well (no one needs .01mm accuracy for furniture making)
-The ability to accept standard sheet sizes without having to saw down the sheet (2'x4' working area was our starting point).. And then to be able to use an auto-feeder, or manually be able to feed the sheet through the machine for the second half of the cuts.
-Some amount of modularity
-ability to accept larger motors (in this case NEMA 34)
-upgrade path (tool changers, spindle control, vacuum, camera system, smart controller, servo and/or feedback system, additional axis)

And of course, this is designed so that it can be offered as a kit, or a stand-alone turn-key product. This means that there's some amount of flexibility in the design to tailor it to your needs. So feel free to send us suggestions. We have a pretty good handle on the first iteration, but that doesn't mean there can't be variant designs for the kits.

R

Drawn by scratch by RW Turner Jr ESQ III, a very talented individual.

Okay, I'm seriously interested. I want to do something that will require some CNC stuff for my car stereo this spring, and it's about time to get cracking.

I've been looking into options to build a CNC, and this looks to be coming up a winner.

Hmmmmm...another hobby rears its ugly head....

Click to view attachment

CURSE YOU BRAINCHIIIIIIIIIIIIIIILD!

Hey Brain, this looks cool from a "wow" standpoint, but help me understand what a CNC will do for me. I have delved into some of the forums and sites, but I come away boggled I guess.

Are the steps:

1) Decide on a project that requires multiple pieces of wood, aluminum, plexiglass, whatever. Or possible scroll designs and fancy cuts.

2) Program the steps necessary to make the cuts from the base wood parts, or to scroll/router the designs.

3) Then feed the various pieces through and assemble

If these are the "CNC for Dummies" steps, then what are examples of what "dummies" have made? For instance, I would love to build a cherry mantle for a fake fireplace in my bedroom. I also would love nice cherry bookshelves in another room. Are there canned programs for the cuts for these type of projects? I worry about getting some big honking cutter, running a lot of expensive wood through only to discover something was off and it has cut the hell of stuff.

Not being critical ... guess I am looking for the "how can this help me" side of things. I am sure people who understand the impact are already fired up. My exposure to this stuff is laser candy bar sculptures and daybooks with my name on it.

@weldonjb: Well, the general idea is that a CNC machine can make accurate cuts/grooves in material following a CAD design. It's advantages are that it makes the lines follow the plan beter than any hand-held tool can, and above all, it's repeatable. As such, once you have a CAD design for, say, a projector case, you can make many of them. Though the repeatability facotry may not be an issue for you, it also means that if you can get the drawing files from someone who has made something you want, you can make an identical unit.

Personally, I'm more interested in the machine's ability to do accurate, repetitive work, such as cutting teeth into a sheet of acryllic so that I can use a nylon driven gear to make parts move. Doing this kind of work by hand is tedious, and near impossible to get right. Add in that I can accurately cut out CAD designs, and the possibilities for projects abount.

The one that I really want to get going, for example, is to rebuild the amplifier rack in the back of my car. I want to have precisely fitted parts which will fit together inside my car which will be strong, so that I can toss stuff in the back of the car, and the amplifiers will be safe. It will also look pretty cool.

There are also numerous projects that I really want to get done that will hugely benefit from the capability of CNC machining. The ones on my "honey do" list currently:

New kitchen cupboard doors. (Without CNC these are likely going to be hardwood borders around flat plywood.)

A drawer set for under my daughter's bed. (I can do this with just a tablesaw and radial arm saw, but getting the CNC to cut the parts for 4 drawers is MUCH easier than doing it myself, and I know that all 4 will be interchangeable, which they will almost certainly NOT be if I do it by hand. In addition, some parts like the drawer slides that I'd normally have to buy I will be able to build.)


And of course, there's my other interests. I mentioned the car audio, but I also want to build a nice display for my DIY home amplifiers, not to mention my projector projects...

I'm waiting on a timeline for these to be available in the LL store, now...

I have always found that if I have a good tool, I find good uses for it. Having the right tools allows for more manufacturing options and a cnc helps with more professional results.

I hope to do some custom inlay on guitars. I have a cheap one to start with.

It would allow you to do stuff very similar to what I did with the light box (and other stuff) on my projector...

First you draw whatever you wanna make on a computard...


When you're satisfied; you lay out all the parts onto a flat plane...


The CNC machin' cuts out your stuff...

If you did everything right, you'll be able to easily put the pieces together into a real-life object...


I had my stuff cut on a CNC water-cutting machine thingy, but the idea is the same.

The possibilities are ENDLESS... Oh man would it be cool to own one of these!

Yea me to - I am setting up my own shop, just for fun, and something like that would be great,

And we're just scratching the surface. With a heater and modeling wax, you could set up a wax deposition printer for creating 3D "lost wax" molds for casting metal, or permanent molds for plaster, rubber and plastics. (I haven't yet done this, but it's simple enough.) One of my favorite sculptors, Bathsheeba Grossman uses this technique; although lately she's been using a "lost wax sintering" process, in which a blast furnace vaporizes the wax, but a bronze alloy power was in the wax and is "sintered" into a solid mass. I have two of her pieces and they are a joy to behold. The 3D crystal "laser disruptive etching" is also fabulous.

LLCNC gantry robots can also plot (draw, cut) very accurately for signage, architecture etc. With 8" of Z, terramapping is possible in foam or other materials. Ice sculpture comes to mind, if the machine is in a walk-in freezer! One of the larger uses of CNC machines is to make more CNC machines. R/C modeling is often cited. With a "probe", sampling of materials and surfaces is possible, and highly detailed maps of surfaces can be created. This is often used for great profit, to duplicate moldings from historic landmarks, or reproduce extremely rare artifacts.

The LLCNC is a highly adaptable, low cost robotics platform that brings rapid prototyping to the public at a price point similar to a good table saw. The uses of the machine are exponentially greater than the saw; Kurzweil is right.

Great questions. One of the more painful aspects of entering this "craft" is to try to understand how you can get info into the robot. Can you make a basic drawing that outlines your cuts? If so, a simple (but great) CAM/cad program like CamBam may be all you need. It has loads of functions for a freeware gcode generator. The functions may seem 'strange' but if you go through the tutorials a couple of times you'll understand. Truly, most if the pain will be figuring out the parametrics for the machine and making it do what you need; but I've already done that based on the open source EMC2 software, meaning you get out of that laborsome chore by installing my freeware LLCNC Parametrics.

I Use Mach3 for the controlling software and V-Carve pro 4.6 for the cam package along with Cut3d and Photo V-Carve.

all are excellent software and easy to use.

Joe

Out of curiosity, what will they be if you use cnc?

Brain, will you have these probes available? Or do you know of a way to make a diy probe? (or of cheap commercial one) A quick search was turning them up at around $2k (ouch).

This makes me wish that I hadn't spent all that money on my pj.

Funny you should mention that...yes I have a few ideas. I've seen ideas regarding DIY probes:

http://www.brusselsprout.org/CNC/1P-Probe/

But the ideas seemed too complicated for the functions required. I envisage a sample probe mounted to an inexpensive potentiometer, or rotary encoder, spring-indexed to a neutral (50/50) position. Sampling and Z control are programmatically linked, allowing the machine to sample and "react" to changes in Z. The LLCNC can then take linear samples of any incremental amount, say .5mm line widths...

3d scan of a penny at .1mm res... impressive.

What are the torque specs on the motors? I noticed the website has a range on each of those models (nema 34, nema 23). Could the x-axis be run with a single, larger motor and a chain/gears? I'm hoping to build/have a 4th axis, so the fewer the motors the better.

Here are a few, they really are easy to build, and you can use mach3 software to drive them.

http://www.indoor.flyer.co.uk/probe.htm

http://www.cnczone.com/forums/showthread.php?t=28308

http://www.cnczone.com/forums/showthread.php?t=22837

Also art is experimenting with a webcam and laser in mach3 to create cloud point files and convert to solid mesh files.

Called Video Probing.

http://www.machsupport.com/forum/index.php...&board=67.0

Joe

Edit: I have read EMC2 has probing also. I have not seen it though.

Hmm, well, I don't think it's a good idea to use anything like a chain...and the extra motor is not a big deal really. EMC2 can control up to hexapod, IIRC. And the controllers we can offer are not expensive.

Nema 34 is around 400oz/in and the 23 we use is about 300oz/in.

Thanks, guys. Helps me understand the uses better.

With the right gantry, for instance, you could load a complete design for an outdoor deck, have the cnc cut all the pieces and angles, drill all the holes for screws, and number letter each end of the boards so that installation would be like putting together a puzzle.

Hole #4 and plank #12 attaches to crossbeam #5 hole #5, etc.

Oooo wait .... GrrAnimal pictures!

Probes,I read recently of guys using aluminum foil and a knitting needle to probe a surface. The idea is the probe is a simple on off switch. They say the difficulty is translating the data to Cloud point, it tends to be time consuming and tedious.

On another note Brain, will you be selling a kit with break out boards wiring and motors? The driver seems like the steepest learning curve, would you agree?

I believe making and testing the controller will be a good jumping off point before actually fabing gantry's tables and stuff.

Yes, we will sell a motor/screw kit, and a driver/computer kit. You can use our embedded computer with our drivers and our parametrics to just start cutting right away. I've already programmed the hard parts, you shouldn't have to as well! The complete system, including computer, will retail for under $2500 if I can help it. You can save some bucks by doing some of the work yourself and buying the pieces as you can afford them. The best deals will come from the pre-buys, since there is ~8 weeks of waiting for the boat, I'm cutting deals to get get the party started! I'm fairly confident that a savvy buyer can complete the machine for under $1k, but a lot more hacking and sourcing falls on you to do so. Naturally, I am here to help.

If the 8 week time frame seems long, no worries. You can start on construction now by grabbing the billet cuts based on the cut-list I will probably be posting tonight (last minute changes, want to be sure there are no errors), and starting the work of precisely drilling and tapping the pieces. You can also assemble the T-Slot base if you don't want to wait for the "Asian extrusion kit".

Yep, you could actually do this with the LLCNC Gantrybot. It is one of the "pass-through" Y roller advantages.

Does the roller work like one of these with a stepper motor?
http://www.grizzly.com/products/1-4-HP-Power-Feeder/G4176

Oh, I see where the evil leads now ...

SJ encourages downstairs renter to lure Red into the "hobby room" again. She then gets her drunk and puts her in the LL cloning CNC, which probes her surface (!), recording the exact dimensions (in and of itself exposing a great mystery for the forum). A logical transform on the data table allows the CNC machine to then chip out a Red mold (not the HG Wells kind), into which is further built the CNC'd aluminum armature and latex layers. Brain then includes said datafile with every purchase. Both become "filthy" rich.

Here we have SJ's definition of a new "hobby".

Not quite. Mine goes the entire X length and has a 'pinch roller' on top to provide positive downward force.

So pretty much the only thing missing at the ~$999 price tag is a controller board? HobbyCNC ($99) or similar will work, I guess. Will you be selling a controller, approximate price? (any eta on when this will all be added to the store?)

Yep, I'm fairly sure you can build it for under $1k. We will sell a PC/controller that is an all-in-one-ready-to-run solution, as well as the parts such as the individual control circuits, power supplies etc.. If you have an old PC, and you can solder up PCBs, you can make your own controllers and save some moolah.

Hmmmmm...somebody read FREEWARE, didnt they?

I have more old PCs than I care to count, and soldering up PCBs is something that I already do as a hobby. This is part of what I find interesting about a DIY CNC, that it meshes with other things that I already enjoy doing.